p.20 #1 · A thinner sensor stack may be possible after all!
Thanks for your replies, gents. I thought it would be the case, but I'm glad you're confirming it. I'm surprised that Sony screwed up so badly on the cover for the A7, that even reflections are better now!
Feb 21, 2015 at 11:09 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.20 #2 · A thinner sensor stack may be possible after all!
mdemeyer wrote:
Agree it changes it. My point is just that thinner changes it less than thicker and the only thing that can be generalized is that, with a thinner stack, performance is closer to the film performance in this regard. Whether that's good or bad for any particular lens takes the level of testing that Lloyd does (very well IMO) that few, if any, others do.
I think we are in violent agreement.
Michael
I think you are in agreement because you are both right. As I understand it the cover glass causes field curvature but especially more in the tangential direction than in the saggital direction leading to increased astigmatism and smearing, but maybe I have that wrong. In any event both smearing and field curvature are issues and It seems the mod helps with both. It will be nice to see whether it completely rectifies the problem or just most so. I think it will be mostly so for lenses designed for film and more so for modern Leica lenses designed for digital Leica M.
p.20 #3 · A thinner sensor stack may be possible after all!
The fact that the thinner sensor glass cover causes legacy lenses to perform better kind of begs the question - why did Sony design the sensor cover to be the thickness it is? And will future Sony sensor covers be the same dimension.
Not trying to generate an argument, but the subject being discussed here is well known to lens and camera designers in the digital age, I would think. So the decision to make the sensor cover the thickness dimension it is would have been discussed beforehand and be based upon a range of parameters. They would have had many examples, from a number of manufacturers, and created their own. Seems like there would be an optimum thickness that the industry would standardize on. Or design them to be interchangeable. Here we see an aftermarket industry developing, did Sony anticipate that?
I hesitate to think that the motivation was to simply sell more lenses, since I think it would have been simple to create interchangeable sensor covers. But maybe Sony was unsure of the market. Being an engineer, I can visualize the discussions going on in the design process, and someone up the corporate ladder saying "We must sell more lenses", and that becomes the dominant design criteria. (May have even been Zeiss saying "we need to sell more lenses, please Sony-san".)
I don't know, but these kinds of corporate decisions are always interesting, like the story behind Apple & Xerox. Anyway, kudos to the guys who have created this product. Lots of interesting glass out there to be explored.
Feb 21, 2015 at 03:18 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.20 #4 · A thinner sensor stack may be possible after all!
If you have followed the thread you will see that the drawbacks of a thinner sensor cover glass have been described. A thinner cover glass will make dust more noticeable, it will make IR contamination more likely, and it will be less durable and susceptible to degradation.
No doubt these factors and perhaps others played a role in Sony's decision to make the sensor glass the thickness it is. Most SLRs seem to have a similar thickness (about 2mm), and Leica is a lone in making their Leica M sensors with thinner glass. That makes sense for Leica because they need compatibility with their film era lenses, but for Sony that is making a new line of lenses I think it makes sense to have the thicker cover glass to avoid the problems of the thinner cover glass. New lenses can be designed to account of the thicker cover glass.
p.20 #5 · A thinner sensor stack may be possible after all!
Yeah, we mustn't forget that Leica hasn't exactly had the best track record the last few years with M sensor issues, sensor glass cracking being one of them.
p.20 #6 · A thinner sensor stack may be possible after all!
To be fair to Leica the issue really only has been the M9 and not the M240. The M240 has evolved and the issue with cracked sensors is not an issue here at least so far.
Having the ability to opt for a thinner filter glass is a great option for those of us here that still have a collection of M glass. There are potential downsides of course but I am sure the techniques will evolve with a number of excellent alternatives. In my discussions with Kolari Vision, there is mention of even better quality thinner filters that can be adapted at a later stage.
With regards to the discussions to the WB, the early FW versions with the M9 were quite problematic, so much so that the preferred RAW developer then was to use C1 Pro as this was then supported by Leica. The color palette in different lighting then with the M9 was difficult and there were numerous threads on this. Later of course Lr was updated supporting Leica and most of issues had been resolved with the M9 FW updates.
Similarly with the M240 the early FW versions the skin tones were very red/orange.
I am waiting for my A7rMod to arrive back in Australia. I realise that I may need a color profile within Lr to suit the "look" I prefer, with the "thin filter". As already suggested WB is a starting point with RAW files and generally the A7r/A7s and A7II are very good in finding the correct WB in difficult lighting.
p.20 #7 · A thinner sensor stack may be possible after all!
douglasf13 wrote:
Yeah, we mustn't forget that Leica hasn't exactly had the best track record the last few years with M sensor issues, sensor glass cracking being one of them.
You mean M9 of course, which also has the best performance ever seen with short register lenses on full frame as regards edges etc, though m240 is not far off and without the CG issues.
The A7.mod is a great improvement. However the M9 and 240 are still way ahead with overall clarity 35 and wider, in my honest opinion. But that comes at quite a price.
I don't have this one quite right as the A7 looses WB bearings in a shot like this, and we know I'm a work in progress editing But you see some colorcast. DSC02516-2 by unoh7, on Flickr
p.20 #8 · A thinner sensor stack may be possible after all!
Hi Charlie,
I do remember if you have no lens correction with the M240, the 21 SEM does behave very similarly in bright light as in your shot above. Leica have done some excellent work in correcting the colour casts.
p.20 #10 · A thinner sensor stack may be possible after all!
uhoh7 wrote:
You mean M9 of course, which also has the best performance ever seen with short register lenses on full frame as regards edges etc, though m240 is not far off and without the CG issues.
The A7.mod is a great improvement. However the M9 and 240 are still way ahead with overall clarity 35 and wider, in my honest opinion. But that comes at quite a price.
So I am happy
Yeah, I'm fully aware of the M9 and its abilities, as I shot one for a couple of years. My point is that there are trade offs and reasons why other manufacturers don't use thin sensor glass, and they're likely not concerned about adapting another brand's glass. We've seen IR contamination, sensor glass cracks and sensor glass corrosion all become issues with M cameras.
p.20 #11 · A thinner sensor stack may be possible after all!
charles.K wrote:
Hi Charlie,
I do remember if you have no lens correction with the M240, the 21 SEM does behave very similarly in bright light as in your shot above. Leica have done some excellent work in correcting the colour casts.
Hi Charles,
Hope you have your camera soon
here another sample to see where the sun was in the shot above: DSC02524 by unoh7, on Flickr
p.20 #12 · A thinner sensor stack may be possible after all!
carlitos wrote:
The fact that the thinner sensor glass cover causes legacy lenses to perform better kind of begs the question - why did Sony design the sensor cover to be the thickness it is?
From everything I've read and tried to extrapolate, I think it goes a bit like this. There is some advantage, however small to the thicker glass. This is why other DSLR manufacturers do it. Sony want to be able to sell sensors to these manufacturers, so they follow suit with their own sensors.
They've chosen to design lenses from scratch and to obtain the performance figures they want they will probably need to go with larger retrofocal designs anyway, so there is no real advantage to Sony in having a thinner stack.
carlitos wrote:
And will future Sony sensor covers be the same dimension.
Yes, for the same reasons.
carlitos wrote:
Seems like there would be an optimum thickness that the industry would standardize on. Or design them to be interchangeable. Here we see an aftermarket industry developing, did Sony anticipate that?
They more or less have, with the exception of Leica. Sony didn't anticipate much demand for the A7 full stop, so I doubt they saw anything like this coming. But this isn't really a "market", this is one company offering a conversion service, for a handful of atypical users. That's nothing compared to the thousands of A7 units shipped.
carlitos wrote:
I hesitate to think that the motivation was to simply sell more lenses, since I think it would have been simple to create interchangeable sensor covers.
Firstly it's not about selling more lenses, as much as it would be Sony going out of their way to make the glass thinner so they could sell fewer lenses. Secondly I don't think many people would be willing to change out their own filter stack or even take it to a company to do it. In many cases the filter stack (or some component of) is bonded to the sensor surface.
p.20 #13 · A thinner sensor stack may be possible after all!
This thread seems to be going in a circular loop. I don't believe there currently exists in the marketplace a thin cover glass based camera that does not suffer some sort of issue. Even the most recent Leica M240 suffers some nasty IR contamination for instance - a deal breaker for some users and uses.
It would be interesting to see a direct comparison between a vanilla A7 and A7 modded with some of the fabrics that are more susceptible to IR contamination (black synthetics are one I believe). It was mentioned earlier that the IR rejection was the same with the mod vs the vanilla A7 but it seems surprising that would be the case (given even Leica still has the issue). That said, I have not seen obvious examples of A7mod posted images that show an issue with IR pollution.
p.20 #14 · A thinner sensor stack may be possible after all!
I think the IR contamination is an overblown issue. The only item I have photographed and shown this contamination was a black Lowe pro camera bag that turned slightly purplish. Apart from this case I have yet to encounter another one, unless I did but didn't notice
p.20 #15 · A thinner sensor stack may be possible after all!
Tariq Gibran wrote:
Even the most recent Leica M240 suffers some nasty IR contamination for instance - a deal breaker for some users and uses.
I have not once encountered any IR issues with the M240. However, I have heard (maybe twice) of MINOR contamination in very rare circumstances. So rare, I would just assume it is a non-issue.
p.20 #17 · A thinner sensor stack may be possible after all!
edwardkaraa wrote:
I think the IR contamination is an overblown issue. The only item I have photographed and shown this contamination was a black Lowe pro camera bag that turned slightly purplish. Apart from this case I have yet to encounter another one, unless I did but didn't notice
It comes down to if the IR contamination that is there is an issue or not for what one shoots. As I posted back on page 6, Ron saw the problem when shooting a wedding with the M240 first hand. A quick search will also turn up other reports. If you are shooting anything that requires critical color accuracy, I think one might have to resort to using an auxiliary IR cut filter on the front of the lens.
p.20 #18 · A thinner sensor stack may be possible after all!
No need for IR filters with M240 as the colors are very accurate. If you try one out for a few days you will soon see it is a non-issue. Using my M8 however was a completely different scenario....
p.20 #19 · A thinner sensor stack may be possible after all!
Gary Clennan wrote:
No need for IR filters with M240 as the colors are very accurate. If you try one out for a few days you will soon see it is a non-issue. Using my M8 however was a completely different scenario....
I think it's just a matter of degrees. The M8 showed extreme contamination. The M240 - as both Ron and now Edward report - still suffers contamination. That seems to be a fact. Whether it matters to one is a different question.
But you see some colorcast.
